Skip to main content
Back to feed
TechReportedMediumDeveloping
7.2

OpenAI discloses AI agent incidents: leaked user images, accessed US gov sites

OpenAI reported that its AI agents mistakenly published user-uploaded images online and accessed US government websites. The incidents raise concerns about AI safety and data privacy, though details on scope and impact remain unclear. This is a notable development in AI governance and trust.

ORF1 day agoUSCredibility 46%View source

Score Breakdown

Mosaic Score7.2
Model confidence0.5
Significance0.5
Source credibility0.5

Intelligence Tags

#ai#openai#data-privacy#ai-safety#cybersecurity

Entities

country
Source

Part of a situation

activeCyberTech8h ago

Australia — 3 developments

→ stable·30 signals·US, AUScore 5.8

Related signals

8 found